EP 0801378 A3 19980930 - Method and apparatus for speech recognition
Title (en)
Method and apparatus for speech recognition
Title (de)
Verfahren und Vorrichtung zur Spracherkennung
Title (fr)
Procédé et dispositif de reconnaissance de la parole
Publication
Application
Priority
US 63187496 A 19960410
Abstract (en)
[origin: EP0801378A2] Methods and apparatus for a language model and language recognition systems are disclosed. The method utilizes a plurality of probabilistic finite state machines having the ability to recognize a pair of sequences, one sequence scanned leftwards, the other scanned rightwards. Each word in the lexicon of the language model is associated with one or more such machines which model the semantic relations between the word and other words. Machine transitions create phrases from a set of word string hypotheses, and incrementally calculate costs related to the probability that such phrases represent the language to be recognized. The cascading lexical head machines utilized in the methods and apparatus capture the structural associations implicit in the hierachical organization of a sentence, resulting in a language model and language recognition systems that combine the lexical sensitivity of N-gram models with the structural properties of dependency grammar. <IMAGE>
IPC 1-7
IPC 8 full level
G10L 15/18 (2006.01); G10L 15/00 (2006.01)
CPC (source: EP US)
G10L 15/193 (2013.01 - EP US); G10L 15/1815 (2013.01 - EP US); G10L 15/197 (2013.01 - EP US)
Citation (search report)
- [A] WO 9416434 A1 19940721 - APPLE COMPUTER [US]
- [A] EP 0238698 A1 19870930 - IBM [US]
- [A] NIJHOLT A: "14. OVERVIEW OF PARALLEL PARSING STRATEGIES", 1991, CURRENT ISSUES IN PARSING TECHNOLOGY, PAGES 207 - 229, MASARU TOMITA,KLUWER ACADEMIC PUBLISHERS, BOSTON/DORDRECHT/LONDON, XP000273591
Designated contracting state (EPC)
DE ES FR GB IT
DOCDB simple family (publication)
EP 0801378 A2 19971015; EP 0801378 A3 19980930; EP 0801378 B1 20021218; CA 2198306 A1 19971011; CA 2198306 C 20010501; DE 69717899 D1 20030130; DE 69717899 T2 20030821; MX 9702521 A 19971031; US 5870706 A 19990209
DOCDB simple family (application)
EP 97302213 A 19970401; CA 2198306 A 19970224; DE 69717899 T 19970401; MX 9702521 A 19970407; US 63187496 A 19960410